K10D sharpness issue!

[mutleybird]

So that takes care of edge sharpening--natural and bright are identical in
characteristic and only different in amount.

3. The texture sharpening characteristics are identical.

I would like to see a more RAW-like jpeg step response

and a soft damped corner on the light side (which is what gives the images their soft appearance.)

Great, careful work Bart. This is far more elegantly put together than my wishful posts for Pentax decision makers, though we are actually agreeing in substance.

Along with keeping this current method of processing that can work for some, there may be a possibility of a Custom Menu option providing acutance/edge sharpening for those that need it because of workflow/time constraints. Maybe, to be more specific, the soft damped corner process would be the thing to be optionally adjusted out to make this a non-issue. Its not worth assuming cut and dry that adjustments by firmware update can't access existing adjustability in PRIME, because its "hardwired" We just can't know. Pentax interviews have mention good adjustability leeway in the PRIME algorithms.

Well done, pity most people after reading this still won't understand the futility of complaining about in-camera jpg sharpness, long live RAW.

Yup, here I went again. I'm keeping my K10D, but just feel this camera is close enough and we don't know enough about PRIME - that it ISN'T futile.

Long Live JPG in DSLRs!

Larry

 

[mutleybird]

So that takes care of edge sharpening--natural and bright are identical in
characteristic and only different in amount.

3. The texture sharpening characteristics are identical.

I would like to see a more RAW-like jpeg step response

and a soft damped corner on the light side (which is what gives the images their soft appearance.)

Great, careful work Bart. This is far more elegantly put together than my wishful posts for Pentax decision makers, though we are actually agreeing in substance.

Along with keeping this current method of processing that can work for some, there may be a possibility of a Custom Menu option providing acutance/edge sharpening for those that need it because of workflow/time constraints. Maybe, to be more specific, the soft damped corner process would be the thing to be optionally adjusted out to make this a non-issue. Its not worth assuming cut and dry that adjustments by firmware update can't access existing adjustability in PRIME, because its "hardwired" We just can't know. Pentax interviews have mention good adjustability leeway in the PRIME algorithms.

Well done, pity most people after reading this still won't understand the futility of complaining about in-camera jpg sharpness, long live RAW.

Yup, here I went again. I'm keeping my K10D, but just feel this camera > is close enough and we don't know enough about PRIME - that it
ISN'T futile.

Long Live JPG in DSLRs!

Larry

 

[mutleybird]

So that takes care of edge sharpening--natural and bright are identical in
characteristic and only different in amount.

3. The texture sharpening characteristics are identical.

I would like to see a more RAW-like jpeg step response

and a soft damped corner on the light side (which is what gives the images their soft appearance.)

Great, careful work Bart. This is far more elegantly put together than my wishful posts for Pentax decision makers, though we are actually agreeing in substance.

Along with keeping this current method of processing that can work for some, there may be a possibility of a Custom Menu option providing acutance/edge sharpening for those that need it because of workflow/time constraints. Maybe, to be more specific, the soft damped corner process would be the thing to be optionally adjusted out to make this a non-issue. Its not worth assuming cut and dry that adjustments by firmware update can't access existing adjustability in PRIME, because its "hardwired" We just can't know. Pentax interviews have mention good adjustability leeway in the PRIME algorithms.

Well done, pity most people after reading this still won't understand
the futility of complaining about in-camera jpg sharpness, long live
RAW.

Yup, here I went again. I'm keeping my K10D, but just feel this camera is close enough and we don't know enough about PRIME - that it
ISN'T futile.

Long Live JPG in DSLRs!

Larry

 

[mutleybird]

Larry

 

[Seb74]

The only reviewer I am aware of who thinks that K10D image
sharpness/detail in default JPEG (without using Bright or
sharpening) isn't up to other 10mp models is Phil Askey of this
site and he says that is indistinguishable difference up to A3
prints.

He sure did a lot of damage with that review, didn't he? The
fallacy that the K10D produces soft images has gotten around.

Joe

Its indeed very easy to spot the difference in the "compared to" section of the review. Noone has to tell anyone anything, its just to look at the pics. Incredibly soft pics indeed, compared to competition, and sharpness ain't supposed to help it either according to review.

I almost never print anything, I use pics on screen, often crop it, so for people like me its very irrelevant if something ain't noticable at A4 or A3 size after print.

 

[distudio]

Its indeed very easy to spot the difference in the "compared to"
section of the review. Noone has to tell anyone anything, its just
to look at the pics. Incredibly soft pics indeed, compared to
competition, and sharpness ain't supposed to help it either
according to review.

I almost never print anything, I use pics on screen, often crop it,
so for people like me its very irrelevant if something ain't
noticable at A4 or A3 size after print.

So how many MP is your screen? Do you regularly crop 50% of the image, not resize and view at 100%, if not the problem that you suggest is moot.

--
Rob

 

[tapir]

The only reviewer I am aware of who thinks that K10D image
sharpness/detail in default JPEG (without using Bright or
sharpening) isn't up to other 10mp models is Phil Askey of this
site and he says that is indistinguishable difference up to A3
prints.

He sure did a lot of damage with that review, didn't he? The
fallacy that the K10D produces soft images has gotten around.

Joe

Its indeed very easy to spot the difference in the "compared to"
section of the review. Noone has to tell anyone anything, its just
to look at the pics. Incredibly soft pics indeed, compared to
competition, and sharpness ain't supposed to help it either
according to review.

Very true, it's easy to see on a 100% crops which incidentally amount to multiple feet sized prints. Moreover a couple of very technical and knowledgeable posters (GordonBGood being one) proved scientficially that bright edge undersharpening algorithm is indeed used by Pentax for both Bright and Natural modes and that in terms of sharpness Bright roughly amounts to Natural with +2 sharpness. Question is how all this is relevant to normal usage of people who never print larger than A3 if at all. I mean phenomena clearly exists but its magnitude should be viewed in a context and not as a major scare.

I almost never print anything, I use pics on screen, often crop it,
so for people like me its very irrelevant if something ain't
noticable at A4 or A3 size after print.

And how much you need to crop to make this noticeable at normal viewing distances.

--
http://www.pbase.com/klopus

 

[jeffkrol]

and does it really matter...........



--
360 minutes from the prime meridian. (-5375min, 3.55sec) 1093' above sea level.

'The exposure meter is calibrated to some clearly defined standards and the user needs to adjust his working method and his subject matter to these values. It does not help to suppose all kinds of assumptions that do not exist.'
Erwin Puts

 

[Seb74]

Its indeed very easy to spot the difference in the "compared to"
section of the review. Noone has to tell anyone anything, its just
to look at the pics. Incredibly soft pics indeed, compared to
competition, and sharpness ain't supposed to help it either
according to review.

I almost never print anything, I use pics on screen, often crop it,
so for people like me its very irrelevant if something ain't
noticable at A4 or A3 size after print.

So how many MP is your screen? Do you regularly crop 50% of the
image, not resize and view at 100%, if not the problem that you
suggest is moot.

--
Rob

I personally often look at pics at full size, or crop and show cool macros and similar to people in 100% crops.
Thats me, I use the pics on my pc, not often at all in prints.

For other people I can see its no issue at all, but sayind that this is no problem at all, and noone should even notice it, that is just wrong and ignorant (or fanboy-attitude).

 

[distudio]

I personally often look at pics at full size, or crop and show cool
macros and similar to people in 100% crops.

Maybe you should choose lenses that make best use of the available imaging area, that's a far more sensible thing to do than suggesting that crops viewed at 100% aren't good. Not that the jpg "un-sharpness" is really an issue at all. If Phil hadn't made a big deal about it we wouldn't be having this conversation, very few other reviews cited this "issue".

BTW Calling somebody a Fanboy for asking you a sensible question just makes you look like an idiot.

--
Rob

 

[Seb74]

I personally often look at pics at full size, or crop and show cool
macros and similar to people in 100% crops.

Maybe you should choose lenses that make best use of the available
imaging area, that's a far more sensible thing to do than
suggesting that crops viewed at 100% aren't good. Not that the jpg
"un-sharpness" is really an issue at all. If Phil hadn't made a big
deal about it we wouldn't be having this conversation, very few
other reviews cited this "issue".

BTW Calling somebody a Fanboy for asking you a sensible question
just makes you look like an idiot.

--
Rob

I'm not the one starting the insults, I'm not that kind of person.

I just said that since the difference were so huge, its pretty ignorant to just....well, ignore it, saying "you shouldn't look at 100% crops, print it or resize it instead". I could as well buy another camera instead that has proper jpeg-managment and gives good pictures straight out of the camera with no need for RAW.

Its just sad, for people that use the camera like I do, when a potentionally very good camera has such an unneccesary software flaw as this

There are always users that totally ignore 100% monitor-views. Pictures can look like s* t, totally noisy at higher ISO's, and someone defends it and says "you damn pixel-peeper, just print it A4 and you wont notice it".

I dont like those people....cameras have all kinds of uses among different people, some dont care if it looks good at 100% on monitor, some do.

I even had a bad pixel on my S3 when I bought it, a white bright pixel almost in the middle of all pics, I was like crazy and went back to the store and the guy said "hey, you gotta realize you wont even see this if you make a A4 or A3 print".

I got a new one though, after some discussion, and it didn't have any sensorfaults luckily.....

 

[Manel Kirchner]

I don't understand so much concern about sharpness when the main
issue with the K10 (as well as with the D80 and other 10 Mg pix
sharing this sensor ) is -and by fa- the CA problem, especially
shooting wide angle.
So to me , this IS a real issue, no sharpness. You can fix
sharpness in excess or a lack of it. But you can spend hours to
hide chromatic aberrations.

Or: one can simply move 2 sliders in Adobe Camera Raw, which is
highly effective and takes even less time than sharpening (if that
is necessary).

I'm curious, do you feel that the 10MP sensors have worse CA than
6MP sensors, assuming the same lens? We are talking about lens CA,
is that right, judging from your comment about wideangle?

As I understand it, that kind of CA shift will not vary optically,
whatever sensor (or film or piece of white paper) is placed at the
imaging plane.

IMO it doesn't quite make sense to then blame the sensor pixel
spacing for a constant physical effect. We could just as well say
that CA is worse when measured in millimetres than when measured in
inches, or that dust specks get magically bigger when they land on
a higher-res sensor! (though 100% crops might well give one that
impression...)

RP

My advice is that you take an 8 million mgapixels Canon Powershot Pro-1 and compare CA with an 4 million megapixels Canon g2 in the same shot. Especially tree branches against a white sky. After that you take the Pro-1 and shoot similar scenes at 28mm and 50 mm. Then please you tell me if I am right or not. You don't need a Canon you can do it with whatever cameras from the 4 and 8 millioon pixels "wannabe- prosumer" -generation

 

[richardplondon]

Of course chromatic aberration is a problem with some cameras. I am unsure whether you are in fact talking about purple fringing instead of CA, from your mention of dark branches against a white sky - but that is also a big problem with many cameras.

My 5Mp Sony V1 suffers badly from PF, and slightly from CA. My 8MP Panasonic LX1 has excellent PF performance in comparison, and CA is well corrected in-camera (I know from comparing against RAW files that the lens does create some CA, but it is scarcely visible in JPGs).

But this does not prove anything about resolution, in fact it may seem to swim against that tide. In fact the two cameras have different generations of sensor, and the more recent one has a much more sophisticated onboard image processor. The effect of these massively swamps the megapixel count as a factor. The sensors are of different physical sizes too. I could go on.

For these reasons the suggested type of test will not prove anything definitive concerning resolution, IMO, leaving aside that different lens and sensor designs are differently prone to CA and PF in the first place.

RP

 

[timo]

Its just sad, for people that use the camera like I do, when a
potentionally very good camera has such an unneccesary software
flaw as this

As has been said a million times, Pentax designed the camera with a certain market in mind. To this end they offered jpg files which were particularly well suited to post processing, which (to a varying degree) is what most serious dslr owners do. The only person who has seen this as a 'flaw' is Uncle Phil - in reality he just disagrees with the decision that Pentax made. If you spend your leisure hours looking at your photos unprocessed at 100% on the screen, then you are not typical. As GordonBGood has pointed out, if you shoot natural-setting at minimum sharpness, and then apply USM, you will get the 'look' you are looking for. As things are, you can print as large as A3 unprocessed and everything is fine; or you can adjust the photo (as most people will do for most dslr pics of all brands), and everything is equally fine. And because of the sharpening default, you will be less likely to end up with the oversharpened monstrosities that populate the photographic world these days. You may wish that Pentax had taken a different route, but all this 'flaw' stuff is not reasonable.
--
tim

 

[Timbert]

Look, this has been covered ad nausium, please stop!

--
Timbert
http://timshouse.zenfolio.com

 

[LimCam]

Hi

Has you question been answered? What is your view on "softness" ?
--
LimCam
Brisbane, Australia

 

[Rupert60]

Man alive! I look at the length of this thread and it reminds me of that song "Who, who, who let the dogs out?" Good work dogs, that guy never even came back, he is still running for his life, with his tail tucked between his legs. Yes, we have guard dogs in Pentax Land. LOL
--
'This is more serious than I thought.....but it is still fun!
http://www.pbase.com/rupertdog Take a look- It's Free!

 

[Seb74]

Its just sad, for people that use the camera like I do, when a
potentionally very good camera has such an unneccesary software
flaw as this

As has been said a million times, Pentax designed the camera with a
certain market in mind. To this end they offered jpg files which
were particularly well suited to post processing, which (to a
varying degree) is what most serious dslr owners do. The only
person who has seen this as a 'flaw' is Uncle Phil - in reality he
just disagrees with the decision that Pentax made. If you spend
your leisure hours looking at your photos unprocessed at 100% on
the screen, then you are not typical. As GordonBGood has pointed
out, if you shoot natural-setting at minimum sharpness, and then
apply USM, you will get the 'look' you are looking for. As things
are, you can print as large as A3 unprocessed and everything is
fine; or you can adjust the photo (as most people will do for most
dslr pics of all brands), and everything is equally fine. And
because of the sharpening default, you will be less likely to end
up with the oversharpened monstrosities that populate the
photographic world these days. You may wish that Pentax had taken a
different route, but all this 'flaw' stuff is not reasonable.
--
tim

You mean all the competition, the highly recommended, world wide critically acclaimed, Nikon D80....well, it, and all other cameras used as comparisons in this review, has unusable oversharpened pictures? No professional user will even look at the dreaded D80 that has so (at least compared to this Pentax) oversharpened jpeg output.

Only Pentax sees the truth, and makes the pics overly soft, in such a intelligent way that not even bringing up incamera sharpness can solve it.

I for sure dont edit every picture I take in Photoshop, thats no fun, I want the ability to get working pictures straight out of cam. Not oversharpened, but not too soft-looking either. I thought the D80 looked soft as default, compared to others, but thats NOTHING compared to Pentax. So if you say Pentax is totally right in what they do, all others must be totally wrong, and I dont buy that....not at all.

Oh well, It's just sad some manufacturers dont let us tweak the settings as we'd like. I'm 100% sure Nikon and Canon and Sony and Olympus lets everyone lower sharpness to get equally soft pics out of their cameras.

No manufacturer should say "you should only use this camera like THIS, so therefore we ignore all other options. Take it or leave it".

 

[lol101]

As you said somewhere else:

I must say I'm getting more and more glad I dont have any DSLR-money > to spend.

I mean come on
The 400D lacks spot metering (of all features out there).
The D80 lacks anti dust sensor.
The K10D lacks sharp jpeg output cause some bad jpeg-processing.

And so it goes, so it goes..."

Keep your S30, save some money and wait for the perfect DSLR to come...

You're entitled to your opinions but so are people actually seiing a benefit from the "bad jpeg processng".

No luck for you, the camera was designed for guys like me who like to apply output-related sharpening as a last step of the image making process.